Comparison of numerical integration methods on highly nonlinear biosystems models

This paper is devoted to the comparison of numerical integration methods for complex dynamical biosystems such as mammalian cell culture process used for monoclonal antibodies production and the baker's yeast process. These bioprocesses are widely used in bio-industry, biochemistry, biology, and medicine. Their mathematical models are highly non-linear and, furthermore, the reaction kinetics is not perfectly known and some of the parameters are not experimentally accessible. Another part of the complexity of these models arises from the high number of state variables that describe their evolution in time. Because some of the parameters of these models need to be estimated on-line, fast and viable numerical integration algorithms need to be employed.